Sharing performance measurement events across domains

Abstract

Network management activities, such as fault analysis and configuration management, are eventually related to changes in network measurements. Some measurement event might be either a trigger or objective of a management activity. We argue that sharing the semantics of performance data across networks provides a basis for more advanced automation. This paper presents an ontology-based system for sharing information about network measurements across network domains. The represented information contains correlations and human-defined mappings between network measurements and the system is based on semantic reasoning that identifies dependencies which arise by combining local and shared information. We demonstrate the usage of the system in a Long Term Evolution (LTE) network domain. Our experiments from an LTE simulator and LTE test network show that a combination of correlations, humandefined mappings, and ontological reasoning produces useful cross-domain information that can be accessed with ontology queries.

Reasoning in agent-based network management

“…It is not feasible to represent the entire state of a mobile network in a single logical model for management purposes. For focused uses such as semantic modelling of configuration management [13], [14], mapping between concepts across domains [12], or analysis of agent coordination [10], classical reasoning is a valid choice. An architecture for knowledge delivery for the purposes of semantic interoperability of autonomic agents has been proposed in [15].…”

“…We shall next describe an example of how ontology constructs defined above ( Figure 2) are used [12]. The approach allows linking effects of operations and requests in the ontology to each other by means of human-defined Key Performance Indicator (KPI) mappings and network measurement effects ( Figure 2).…”

“…Thus, both of the functions would be valid solutions to a request having similar effect as an objective. More details about the effect modelling is defined in [12].…”

“…As the Figure 2 depicts, the reasoner uses effects of agent operations, incoming requests, and network measurements. Using these as input, reasoner first performs inference to find relationships between effects [12]. Next, reasoner maps operations to requested effects with Equation 1.…”

“…Network is represented in the test system by an internal LTE simulator with configuration and Performance Management (PM) Application Programming Interfaces (APIs). The simulator supports research of agents monitoring LTE KPIs such as Channel Quality Indicator (CQI) and Radio Link Failure (RLF) statistics and performing configurations to transceiver (TRX) power and antenna tilt (RET) parameters, for example [12]. We used Capacity and Coverage Optimization (CCO) SON function based on fixed rule set in the test system.…”